Article pubs.acs.org/est
Aerobic Soil Biotransformation of 6:2 Fluorotelomer Iodide Ting Ruan,† Bogdan Szostek,‡ Patrick W. Folsom,‡ Barry W. Wolstenholme,‡ Runzeng Liu,† Jiyan Liu,† Guibin Jiang,*,† Ning Wang,*,‡ and Robert C. Buck‡ †
State Key Laboratory of Environmental Chemistry and Ecotoxicology, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Post Office Box 2871, Beijing 100085, People’s Republic of China ‡ DuPont Haskell Global Centers for Health and Environmental Sciences, E. I. du Pont de Nemours and Company, Incorporated, Glasgow 300, Post Office Box 6300, Newark, Delaware 19714-6300, United States S Supporting Information *
ABSTRACT: 6:2 FTI [F(CF2)6CH2CH2I] is a principal industrial raw material used to manufacture 6:2 FTOH [F(CF2)6CH2CH2OH] and 6:2 FTOH-based products and could enter aerobic environments from possible industrial emissions where it is manufactured. This is the first study to assess 6:2 FTI aerobic soil biotransformation, quantify transformation products, and elucidate its biotransformation pathways. 6:2 FTI biotransformation led to 6:2 FTOH as a key intermediate, which was subsequently biotransformed to other significant transformation products, including PFPeA [F(CF 2 ) 4 COOH, 20 mol % at day 91], 5:3 acid [F(CF2)5CH2CH2COOH, 16 mol %], PFHxA [F(CF2)5COOH, 3.8 mol %], and 4:3 acid [F(CF2)4CH2CH2COOH, 3.0 mol %]. 6:2 FTI biotransformation also led to a significant level of PFHpA [F(CF2)6COOH, 16 mol % at day 91], perhaps via another putative intermediate, 6:2 FTUI [F(CF2)6CHCHI], whose molecular identity and further biotransformation were not verified because of the lack of an authentic standard. Total recovery of the aforementioned per- and polyfluorocarboxylates accounted for 59 mol % of initially applied 6:2 FTI by day 91, in comparison to 56 mol % when soil was dosed with 6:2 FTOH, which did not lead to PFHpA. Thus, were 6:2 FTI to be released from its manufacture and undergo soil microbial biotransformation, it could form PFPeA, PFHpA, PFHxA, 5:3 acid, and 4:3 acid in the environment.
■
INTRODUCTION Perfluorocarboxylates [e.g., perfluorooctanoic acid (PFOA)] and sulfonates [e.g., perfluorooctane sulfonic acid (PFOS)] were detected in various environmental compartments1−4 and biological species, including humans,4−7 because of their widespread consumer and industrial use and subsequent release to the environment, including surfactants, lubricants, paints, and coatings over the past 60 years.8 The toxicity of PFOS and PFOA in laboratory animals and humans was reviewed recently.4 Because of their persistence and potential toxicity, these perfluoroalkyl substances (PFASs) have been subjected to increased public scrutiny in recent years. The environmental loading of PFASs is generally related to direct manufacturing emissions and secondary exposure from use and disposal of consumer products, as well as indirect transformation of precursors.9,10 Such precursors include perfluoroalkane sulfonyl-based (e.g., N-ethyl perfluorooctanesulfonamide) and fluorotelomer-based [e.g., fluorotelomer alcohol (FTOH)] products.9,10 Polyfluorinated fluorotelomer iodides (FTIs) are key industrial intermediates used to synthesize FTOHs and other FTOH-based products for consumer and industrial applications.11,12 FTOH-based consumer products contained less than © 2013 American Chemical Society
0.05% residual 8:2 FTI [F(CF2)8CH2CH2I] and 6:2 FTI [F(CF2)6CH2CH2I],13,14 which may be subjected to abiotic and microbial biodegradation when the products are disposed to the environment at the end of their life cycle.12,15−17 6:2 FTI is the principal industrial raw material used to manufacture 6:2 FTOH and 6:2 FTOH-based products by major global manufacturers.11 6:2 FTI manufacturing sites could be potential sources for 6:2 FTI to be released to the environment during production. For example, 6:2 FTI levels reported at a manufacturer site and in the vicinity ranged from 1.4 to 368 ng m−3 air in the atmosphere and 17 ng kg−1 in surface soil, while 8:2 FTI ranged from 1.3 to 1320 ng m−3 and from 145 to 499 ng kg−1, respectively.12 6:2 FTI has a predicted vapor pressure of 386 Pa and log Koc of 5.6, compared to 129 Pa and 3.6 for 6:2 FTOH, respectively.18 6:2 FTI released to the environment via potential industrial emission may undergo abiotic degradation in the atmosphere and biotransformation in surface soil. Received: Revised: Accepted: Published: 11504
April 24, 2013 September 5, 2013 September 10, 2013 September 10, 2013 dx.doi.org/10.1021/es4018128 | Environ. Sci. Technol. 2013, 47, 11504−11511
Environmental Science & Technology
■
No information is currently available on 6:2 FTI biotransformation potential in the environment. Therefore, it is not known whether or not 6:2 FTI will be transformed to the same perfluorocarboxylic acids (PFCAs) and polyfluoro carboxylic acids via similar biotransformation pathways as described for 6:2 FTOH [F(CF2)6CH2CH2OH].19−21 The terminal iodide of the 6:2 FTI molecule is a good leaving group with relatively strong nucleophilicity compared to hydroxyl and other functional groups.22 This unique property may facilitate 6:2 FTI conversion via biotransformation pathway(s) otherwise not available to 6:2 FTOH and other 6:2 FTOH-based analogues. Consequentially, 6:2 FTI may be biotransformed by microbes in the environment to form some stable product(s) with higher molar yields compared to 6:2 FTOH and 6:2 FTOH-based analogues. 6:2 FTOH aerobic biotransformation in soil,19 sediment,20 and activated sludge21 generated two types of major transformation products: PFCAs [e.g., perfluorobutanoic acid (PFBA), perfluoropentanoic acid (PFPeA), and perfluorohexanoic acid (PFHxA)] and x:3 acids ([F(CF2)xCH2CH2COOH, where x = 3, 4, or 5]. Perfluoroheptanoic acid (PFHpA) was not observed in any of these aerobic studies. 6:2 FTOH anaerobic biotransformation in digester sludge formed mostly polyfluorocarboxylic acids and low levels of PFHxA (≤0.4 mol %).23 Aerobic biotransformation of 6:2 fluorotelomer analogues, such as 6:2 fluorotelomer sulfonate [6:2 FTSA, F(CF2)6CH2CH2SO3−] and 6:2 fluorotelomer ethoxylate, in activated sludge or sludge effluent generally followed similar degradation pathways as 6:2 FTOH, which was either bypassed or initially formed and subsequently converted to PFPeA, PFHxA, and 5:3 acid.24,25 Aerobic biotransformation of 6:2 fluorotelomer phosphates in activated sludge with inorganic phosphate being removed also formed 6:2 FTOH, PFPeA, PFHxA, and 5:3 acid, as well as 1−1.3 mol % PFHpA.26 It is not clear whether these low levels of PFHpA observed were from 6:2 fluorotelomer monophosphate [6:2 monoPAP, F(CF2)6CH2CH2H2PO4] and 6:2 fluorotelomer diphosphate [6:2 diPAP, [F(CF2)6CH2CH2]2HPO4] or impurity biodegradation, because no information was presented regarding the purity of the two test materials.26 Soil biotransformation of 8:2 fluorotelomer stearate monoester27 and citrate triester28 also generated 0.4−0.9% PFHpA. PFHpA, along with other PFCAs detected in the environment and biota, may come from direct emission and precursor degradation. PFHpA ranged from 1.7 to 43 ng L−1 in activated sludge of WWTPs, up to 7.5 ng g−1 in river sediment, and up to 30 ng L−1 in surface water.29 PFHpA in human blood ranged from nondetectable to 0.4 ng mL−1 in U.S. populations surveyed between 1999 and 2011.30 Cytotoxicity of PFHpA in a marine bacterium, rat cell lines, and human liver cells ranked between PFHxA and PFOA.31,32 The blood elimination half-life is less than 30 days for PFHxA and approximately 1.5−4.5 months for PFHpA in ski-wax workers exposed to high levels of FTOHs via inhalation.33 The goals of this study were to investigate 6:2 FTI aerobic soil biotransformation potential, to quantify biotransformation products, and to elucidate 6:2 FTI biotransformation pathways. The outcomes of this study can help discern the sources of poly- and perfluoro carboxylates detected in the environment.
Article
MATERIALS AND METHODS
Materials. Chemical names, acronyms, and molecular structures of the poly- and perfluoroalkyl substances described in this paper are shown in Table SI-1 of the Supporting Information. 6:2 FTI (purity of 98%) was obtained from Alfa Aesar (WardHill, MA). All other chemicals mentioned below have 96% or greater purity as previously described.19 6:2 FTOH, PFBA, PFPeA, PFHxA, and PFHpA were purchased from Sigma-Aldrich (St. Louis, MO). 5:2 ketone [F(CF2)5C(O)CH3] was obtained from TCI America (Portland, OR). Additional unlabeled fluorinated standards {6:2 FTCA [F(CF2)6CH2COOH], 6:2 FTUCA [F(CF2)5CFCHCOOH], 5:2 sFTOH [F(CF2)5CH(OH)CH3], 5:3 acid, 4:3 acid, 5:3 Uacid [F(CF2)5CHCHCOOH], and α-OH 5:3 acid [F(CF2)5CH2CH(OH)COOH]} used for quantitative analysis were from DuPont (Wilmington, DE). [1,1,2,2-D;3-13C] 6:2 FTOH [F(CF2)513CF2CD2CD2OH] (DuPont, Wilmington, DE) and [1,2-13C] PFHxA [F(CF2)413CF213COOH] (Wellington Laboratories, Ontario, Canada) were used as quantitation internal standards in liquid chromatography/tandem mass spectrometry (LC/MS/MS) analysis. C18 cartridges (600 mg of sorbent) were purchased from Alltech (Deerfield, IL). Acetonitrile was LC/MS-grade, and all other solvents were high-performance liquid chromatography (HPLC)-grade. Deionized water (18 MΩ) for biodegradation studies was generated by a Barnstead E-Pure system. Experimental System. The Sassafras soil (Ultisol) used to study 6:2 and 8:2 FTOH aerobic biotransformation19,34,35 was also selected as the microcosm to study 6:2 FTI aerobic biodegradation. The soil was collected from the same site in an undisturbed forested area for more than 40 years in Newark, DE in March, 2012, was sieved immediately upon collection via a 2 mm sieve, and was stored at 4 °C until use. The soil consisted of 52% sand, 34% silt, and 14% clay, with 3.8% organic matter and pH of approximately 5.8. The initial gravimetric moisture content of approximately 19% in live soil and 10% in sterile soil was adjusted to approximately 50% of the maximum soil water holding capacity by adding deionized water and equilibrating for 2 days before use. Crimp-sealed glass serum bottles (120 mL volume) with butyl rubber stoppers were used as the test vessels. Soil was also sterilized by Cobalt-60 γ-ray irradiation. Triplicate vessels for dosed live soil (live) and sterile-soil negative control (sterile) and duplicate live soil blank matrix (matrix, dosed with carrier solvent only) were individually prepared to cover each sampling time (days 0, 1, 3, 9, 14, 28, 58, and 91). Sample bottles containing 8.3 g of dry weight equivalent soil were incubated for 2 days before being dosed with 6:2 FTI or pure ethanol. For the live group, 10 μL of 6:2 FTI stock solution (4.0 mg mL−1 made in pure ethanol) was dosed into each bottle to a final concentration of approximately 4.0 μg g−1 of soil (8.4 nmol g−1 of soil). For the live matrix control group, only 10 μL of ethanol was dosed into each sample bottle. Prior to adding 6:2 FTI stock solution to sterilized soil samples, triple antibiotics (kanamycin, chloramphenicol, and cycloheximide) were dosed into each bottle to a final concentration of 100 mg kg−1 of dry soil to retard potential microbe growth. After 6:2 FTI or ethanol dosing, the soil in each sample bottle was thoroughly mixed with a metal spatula. After each test vessel was filled with appropriate test media, each bottle was sealed with a sterilized butyl rubber septum/ aluminum cap. One acetonitrile pre-rinsed C18 cartridge was 11505
dx.doi.org/10.1021/es4018128 | Environ. Sci. Technol. 2013, 47, 11504−11511
Environmental Science & Technology
Article
ether (MTBE), which allowed for multiple analysis. The remaining soil in a bottle, after being saturated with deionized water (∼12 mL), was extracted with 10 mL of MTBE instead of 30 mL of acetonitrile. Detailed information is summarized in the Supporting Information. A Waters gas chromatography time-of-flight mass spectrometer equipped with an Agilent 6890N gas chromatograph system (GC−TOF MS) was employed in the full-scan mode (25−800 amu) to detect and elucidate structures of volatile biotransformation products collected with the SPME fibers or present in the MTBE extracts of C18 cartridges. For live and matrix control soil extracts, a LTQ Orbitrap MS (Thermo Fisher Scientific, Inc., Waltham, MA) with electrospray negative ionization was used. Full-scan (m/z 100−1000) mass data with a resolution of 30 000 were acquired to identify potential biotransformation products. Mass defect filtration (m/z 100− 500, from −50 to +10 mDa) was applied to the data to eliminate the signals that are not pertinent to polyfluorinated transformation products. Molecular structure elucidation was based on accurate mass measurement of deprotonated molecule ions and corresponding product ion spectra. Neat 6:2 FTI used in this study was analyzed on the GC−TOF MS system equipped with a MS detector and a flame ionization detector (FID) (see Table SI-5 of the Supporting Information) to identify the impurities with MS detector and quantify the impurities with FID by calculating the impurity peak area percentage from the FID chromatograms. The putatively identified impurities in the neat 6:2 FTI were examined as potential precursors to PFHpA. Three impurities with unconfirmed chemical structures (because of the lack of authentic standards), which may be potential PFHpA precursors, were found in the neat 6:2 FTI: F(CF2)6CH CHI (6:2 FTUI, 0.021%; see Figure SI-1 of the Supporting Information), F(CF2)6CH(OH)CH2I (2-OH 6:2 FTI, 0.11%), and F(CF2)6CH2I (6:1 FTI, 0.04%).
used as the conduit through an 18-gauge needle pierced through the septum to the headspace for aeration and also for capturing parent compound and potential volatile transformation products.20 All bottles were incubated statically in the dark at room temperature (22 ± 3 °C) prior to sample processing and analysis. The headspace oxygen concentration in each matrix control bottle dosed with only pure ethanol was monitored periodically in situ with a 23-gauge needle oxygen probe (model 905, Quantek Instruments, Grafton, MA) to approximate the O2 content in live soil groups. The headspace of the live and matrix control groups were purged with approximately 3 L of ambient air when the oxygen content was near or below 10%. The oxygen level at 10% or above in the headspace of the live soil bottles would ensure an aerobic environment during 6:2 FTI biotransformation because previous work36 showed that 8:2 FTOH was biotransformed by activated sludge when the headspace oxygen level was above 4% in closed bottles. Sample Pretreatment and Quantitative Analysis. After the headspace was purged with about 3 L of air through the C18 cartridge, the sample bottles were sacrificed for extraction with acetonitrile. The sample extraction procedures including C18 cartridge elution, septum and soil extractions and EnviCarb cleanup of soil extract to recover soil-bound 5:3 acid in Sassafras soil and instrument quantification measurements were similar to those previously described,19,20,35 with slight modifications. Detailed information is given in the Supporting Information. The LC/MS/MS quantitative analysis was performed on a Waters Quattro Micro triple-quadruple mass spectrometry system interfaced with a Waters 2795 HPLC module (Waters, Milford, MA). Electrospray ionization was operated in negative mode, and multiple reaction monitoring (MRM) was used for 6:2 FTOH, 5:2 sFTOH, 5:2 ketone, and all other potential polyfluorinated acid transformation products. An Agilent 6890 GC/5973 MS system (Santa Clara, CA) was also used for quantifying 6:2 FTI in all test vials. Electron impact (EI) ionization positive mode was used, and selective ion monitoring (SIM) helped increase sensitivity. Because no isotope-labeled internal standards were available for FTIs, external standard calibration curves were used to perform quantification. Structural Elucidation of Biotransformation Products. To identify biotransformation intermediates and establish 6:2 FTI biotransformation pathways, live and matrix groups of test vessels with the incubation period of days 0, 3, 9, 14, and 28 were also prepared. The microcosm setup in live soil group bottles was similar to the 91 day exposure experiment, except that 12.5-fold more 6:2 FTI (500 μg in 125 μL of pure ethanol) was added to 10 g of soil inside the glass serum bottle (50 μg g−1 of soil or 105 nmol g−1 of soil) to enhance detection of transformation products with low molar yields. Collection of volatile transformation products accumulated in the headspace of the serum bottles was performed by either solid-phase microextraction (SPME) or trapping of the headspace onto a C18 cartridge. SPME sampling of the headspace was performed with a polydimethylsiloxane (PDMS) fiber equilibrated with the headspace for 40 min, which allowed for a single analysis. SPME sampling was used to maximize the detection potential of volatile transformation products because of high concentration factors and solventless sampling to avoid interfering solvent peaks. Using another set of bottles, the headspace was purged with about 3 L of ambient air into the C18 cartridge. The C18 cartridge was eluted with 3 mL of methyl tert-butyl
■
RESULTS AND DISCUSSION Experimental System. Figure 1 shows the trend of oxygen levels in the headspace of the live matrix control groups dosed with ethanol. When 10 μL of pure ethanol was added to the live sample bottles, the oxygen level decreased to 18.3% at day 3 and then gradually increased to levels near that in ambient air
Figure 1. Oxygen levels in the headspace of the live matrix control soil bottles dosed with 10 or 125 μL of pure ethanol. The arrows indicate that the headspace was purged with approximately 2 L of ambient air after O2 measurement. 11506
dx.doi.org/10.1021/es4018128 | Environ. Sci. Technol. 2013, 47, 11504−11511
Environmental Science & Technology
Article
by day 14. This suggests that most of the ethanol was mineralized by day 14 and ambient air was in exchange with the headspace through C18 cartridge conduit. When 125 μL of ethanol was added to the live sample bottles, more oxygen was consumed and the oxygen level decreased to 11.4% by day 3. After headspace was purged, the oxygen level again decreased to 10.8% at day 7 and then back to 16−18% thereafter, suggesting that some remaining ethanol was still consuming oxygen after day 14. For both ethanol levels, the experimental system was aerobic during 6:2 FTI biotransformation. It is not clear whether the addition of ethanol to the soil experimental system can help enhance 6:2 FTI biotransformation. It is possible that ethanol as an additional organic carbon source in the experimental system may provide metabolic energy and increase microbial populations before it was mineralized. However, it is not known whether increased microbial populations are positively correlated to 6:2 FTI biotransformation potential in soil. Previous work36 showed that number 3 carbon of [3- 14 C] 8:2 FTOH [F(CF2)714CF2CH2CH2OH] was mineralized to yield about 12% 14CO2 by activated sludge with periodic addition of ethanol. Because microbial populations in activated sludge may be different from that in soil, further work is needed to determine if the addition of ethanol may promote soil microbes to transform 6:2 FTI and other 6:2 FTOH-related analogues. Mass Balance. The partitioning of 6:2 FTI into the headspace (gas phase) after initial dosing to the soil compartment occurred simultaneously with the aerobic biotransformation processes in the live incubation system (Figure 2). 6:2 FTI soil biotransformation gradually reduced its
Figure 2. Partitioning of 6:2 FTI into the headspace (gas phase, recovered from C18 cartridges and septa) and soil phase (recovered from first and second soil extracts) of the live and sterile soil sample bottles.
levels in both the headspace and soil (Figure 2). 6:2 FTI recovered from the gas phase and soil was 7.1 mol % of initially applied 6:2 FTI in the live test vessels by day 91 (Figures 2 and 3A). In sterile control vessels, the majority of 6:2 FTI was detected in the gas phase from days 3 to 91 (Figure 2). Because 90−112 mol % of initially applied 6:2 FTI was recovered in sterile control vessels between days 3 and 59 (Figure 3A), this indicates that C18 cartridges were effective at capturing volatilized 6:2 FTI. C18 cartridge efficiency to capture 6:2 FTI may be decreased after 59 days because 70 mol % of initially applied 6:2 FTI was recovered in sterile control vessels at day 91 (Figure 3A).
Figure 3. Total molar percentage recovery (mass balance) of 6:2 FTI and quantified transformation products in the incubation system dosed with 40 μg of 6:2 FTI in 10 μL of ethanol per sample bottle containing 10 g of soil. (A) Total mass balance of dosed live soil group (n = 3) and sterile soil control (n = 3) and time trend of 6:2 FTI in live soil. (B) Molar yields of observed PFCAs (i.e., PFBA, PFPeA, PFHxA, and PFHpA) and x:3 acids (5:3 and 4:3 acids). (C) Molar yields of observed intermediate biotransformation products. 11507
dx.doi.org/10.1021/es4018128 | Environ. Sci. Technol. 2013, 47, 11504−11511
Environmental Science & Technology
Article
with significant molar yields by microbes in the environment. Other analogues, including 6:2 FTOH itself, were either not converted to PFHpA or with low conversion yields (Table 2 and references therein). Similarly, 8:2 FTOH34,36,37 and its analogues27,28 were also not transformed to PFNA by microbes (Table 2), and thus, 8:2 FTOH-based products are unlikely a significant source of PFNA detected in the environment via microbial biotransformation. However, it is possible that 8:2 FTI could be transformed to PFNA via a similar microbial pathway for 6:2 FTI. Future experimental work is needed to confirm this possibility. 5:3 acid reached peak levels at day 14 at 22 mol % and then decreased to 16 mol % by day 91 (Figure 3B and Table 1), comparable to the observed 5:3 acid level in soil dosed with 6:2 FTOH,19 and 4:3 acid accounted for 3.0 mol % at day 91. The decreased 5:3 acid level after day 14 may reflect increased tendency for 5:3 acid to be bound to soil and was therefore more difficult to recover even with solvent extraction and postprocessing with base extraction and EnviCarb cleanup procedures.20,35 Transient polyfluoroalkyl acids, such as 6:2 FTCA, 6:2 FTUCA, and 5:3 Uacid, were detected at low concentrations during 6:2 FTI biotransformation in soil (Figure 3C), likely because of their fast further degradation to downstream biotransformation products, as was observed in 6:2 FTOHdosed soil.19 Low levels of 6:2 FTOH (
